When the Audience Joins the Investigation

On Aug. 31, a Las Vegas jury did something that had seemed increasingly unlikely with each passing year: it returned a murder conviction in the 1996 killing of Tupac Shakur.

Duane "Keffe D" Davis was found guilty of first-degree murder for orchestrating the drive-by shooting that killed the 25-year-old rapper nearly 30 years ago. Prosecutors did not argue that Davis pulled the trigger. They argued that he supplied the gun and helped organize the retaliatory attack. The jury deliberated for roughly three hours. Sentencing is scheduled for Oct. 13, and Davis has said he plans to appeal.¹²

The verdict is tempting material for the modern true-crime machine: one of the most famous cold cases in American popular culture finally reaches a courtroom, decades of theories collapse toward a legal conclusion, and a mystery that filled documentaries, message boards, books, interviews and late-night arguments suddenly acquires a verdict form.

But the Tupac case is not an internet-sleuth success story. That distinction matters.

The prosecution's case relied heavily on Davis's own public and recorded accounts of the killing, including interviews and his 2019 memoir. In a strange reversal, years of media afterlife around the case became part of the evidence used to prosecute it.¹² The lesson is not that Reddit solved Tupac. It is that a cold case can now remain so culturally active, so repeatedly discussed and re-narrated, that the public version of the case starts feeding back into the legal one.

True crime used to arrive after the investigation. Increasingly, it lives inside it.

True crime stopped being passive

People have consumed stories about real violence for centuries. What changed in the last decade was not simply appetite. It was participation.

The podcast boom gave audiences hours instead of minutes, documents instead of summaries and hosts who often narrated the act of investigation itself. Pew Research Center found that true crime was the single most common subject among top-ranked U.S. podcasts in its 2023 audit, accounting for 24% of the shows studied. Among Americans who had listened to a podcast in the previous year, 34% said they regularly listened to true-crime programs.³⁴

That format changed the relationship between story and audience. A listener could finish an episode, search a name, pull an old newspaper clipping, compare a map, join a subreddit and be arguing about a witness statement before the next episode dropped. The documentary audience became a research pool.

Criminologists have been describing this as "websleuthing" for years: amateur investigation in networked spaces where people collectively assess evidence, construct theories and grant one another informal investigative authority.⁵ More recent research on online crime discussions shows how that authority can emerge socially. Posters demonstrate familiarity with a case, other users reward convincing interpretations, and a person who began as another commenter can gradually be treated as someone who "knows" the file.⁶

That process can produce useful attention. It can also produce confidence faster than evidence.

Sometimes the crowd really does find something

The strongest defense of participatory true crime is that there are cases where public attention has done measurable good.

In 2015, a man who had been known for two decades only as "Grateful Doe" was identified as 19-year-old Jason Callahan. He had died in a 1995 Virginia car crash while following the Grateful Dead. Volunteers on Websleuths, Reddit, Facebook and other platforms circulated his reconstruction and details of the case until people who remembered a young man named Jason came forward. The publicity eventually reached Callahan's mother, who filed a missing-person report, and DNA confirmed the identification.⁷

The important part of that story is not that strangers on the internet declared an answer and were proved right. It is that they created visibility, generated a plausible lead and moved it into a system capable of verification.

That is also how the Doe Network describes its own role. The volunteer nonprofit, active since 2001, says it works to expose missing and unidentified-person cases, develop credible potential matches and send relevant information to investigating agencies. Its site explicitly says: "We are not private detectives."⁸

There is plenty of legitimate work for that kind of public attention. NamUs reported more than 15,000 active unidentified-decedent cases as of June 1, 2026, and roughly 90,000 active missing-person records may exist nationally on any given day. Since the federal NamUs program began, more than 61,000 missing, unidentified and unclaimed-person cases entered into the system have been resolved, with more than 11,300 identifications indicated as involving DNA, forensic genetic genealogy, dental, fingerprint, anthropological or radiographic comparison.⁹¹⁰

The internet did not create the cold-case backlog. It created millions of additional eyes capable of seeing pieces of it.

A podcast can become investigative infrastructure

The Kristin Smart case offers a second, more professionalized model.

Smart disappeared from California Polytechnic State University in 1996. In 2019, musician and podcaster Chris Lambert launched Your Own Backyard, a deeply reported series that revisited the case, interviewed witnesses and assembled old information into a form a new audience could follow. San Luis Obispo County Sheriff Ian Parkinson later credited the podcast with drawing out witnesses and generating information useful to investigators. Paul Flores was convicted of Smart's murder in 2022 and sentenced in 2023.¹¹

The distinction from casual sleuthing is important. Lambert was not simply posting a suspect's photograph and asking thousands of people to decide whether the face looked guilty. He reported. He interviewed. He checked documents. He maintained communication with authorities. The podcast created pressure and surfaced witnesses, but the criminal case still had to be built under evidentiary rules.

That is what productive public participation looks like at its best: attention expands the investigation without pretending to replace it.

Then there is Boston

The counterexample is so well known it should probably be taught as basic internet literacy.

After the 2013 Boston Marathon bombing, Reddit users created a forum devoted to identifying the bombers from photographs and fragments of information. Among the innocent people swept into the speculation was Sunil Tripathi, a 22-year-old Brown University student who had already been missing for weeks. Users circulated his name and photograph as a possible suspect. Rumors from police-scanner chatter were repeated as fact. Conventional media amplified some of the speculation. Tripathi was not involved.¹²¹³

Reddit later apologized for what it called online witch hunts and dangerous speculation. Tripathi's family had already been searching for their missing son while strangers flooded the internet with accusations attached to his name. His body was recovered days later in Rhode Island.¹²¹³

The Boston failure is often described as a warning about crowdsourcing. More precisely, it is a warning about collapsing the distance between a lead and a conclusion.

Police can receive a tip and investigate it privately. A journalist can receive an allegation and withhold it until it is corroborated. An online crowd can receive the same allegation and turn it into a search result attached permanently to someone's name in minutes.

That difference is not procedural trivia. It is the ethical core of the problem.

The family is not a supporting character

True crime's largest ethical weakness may be built into its grammar. The genre needs a story. Families are living through an event.

A 2025 study by Kelli S. Boling and Danielle C. Slakoff interviewed 20 people who described themselves as co-victims - friends or relatives of missing people, victims or perpetrators whose cases had been featured in true-crime productions. They identified five recurring harms: inaccuracies, sensationalization, loss of privacy, uncomfortable interactions with consumers including online trolls, and lack of control over how their stories were produced or edited.¹⁴

That list should make almost every casual true-crime convention feel slightly less casual.

A listener can stop an episode. A family cannot stop being related to the dead person. A creator can move on to next week's case. A sister, parent, child or spouse may spend years dealing with strangers who now believe they understand the family because they watched six hours about it.

The genre becomes most ethically unstable when participation is mistaken for standing. Knowing the evidence does not make someone part of the family. Caring about the case does not create a right to contact them. A public court record does not make every private detail morally available for content.

This is where journalism's old rules remain surprisingly modern. The Society of Professional Journalists' ethics code tells reporters to verify before publishing, provide context, seek responses from people accused of wrongdoing, balance the public's need for information against harm, avoid pandering to lurid curiosity and consider the lasting consequences of publication. Its 2026 revision project retains those principles while placing even greater emphasis on verification and responsibility in a changed information environment.¹⁵¹⁶

True-crime creators are not all journalists. The people consuming their work certainly are not. The ethical problem is that the consequences of publication do not care what title anyone uses.

Cold cases are especially vulnerable to fandom logic

A cold case creates an almost perfect environment for participatory certainty.

The record is incomplete by definition. Witnesses are dead or aging. Physical evidence may be lost. Old police practices can look bizarre through modern eyes. Newspaper archives contain contradictions. A person who was once a suspect may have been cleared through information never made public. Gaps invite interpretation, and the internet is extraordinarily good at filling empty space with narrative.

Research on true-crime audiences complicates the caricature that everyone watching is simply hungry for gore. A 2026 study by sociologist Christine Zozula found that consumers and creators often described justice-oriented, therapeutic and community motives for their engagement. Zozula uses the idea of "repair work" to describe how some participants understand true crime as a way of making sense of harm or helping themselves and others.¹⁷

Those motives can be sincere and still produce bad outcomes.

A person searching for justice is not automatically good at evaluating evidence. A community built around helping can still reward the member who produces the most dramatic theory. An unsolved case can develop lore: favored suspects, canonical timelines, disputed clues elevated into sacred objects. The longer the mystery lasts, the easier it becomes to forget that uncertainty is not a plot device. It is the actual state of the evidence.

The new tool kit is real

None of this means the public should stop looking.

The digital tool kit available to cold-case investigators, journalists and responsible volunteers is extraordinary: searchable court dockets, newspaper archives, digitized yearbooks, satellite imagery, public-record databases, NamUs, social-media archives, genealogy tools and open-source verification techniques that were once available only to institutions.

Open-source investigation itself has matured. Research published in 2025 described techniques once associated with informal internet sleuthing becoming institutionalized inside journalism, with organizations such as Bellingcat and Forensic Architecture demonstrating methods for verifying publicly available images, videos, satellite data and records.¹⁸

The important word there is verifying.

The evolution of true crime has produced a spectrum. At one end is disciplined open-source investigation: preserve provenance, document uncertainty, corroborate independently and publish only what the evidence supports. At the other is fandom with subpoenas imagined into it: screenshots, vibes, facial resemblance and the belief that spending 80 hours in a forum has conferred jurisdiction.

They can look similar from a distance but they are not the same practice.

A tip is not a verdict

The internet is most useful to a cold case when it knows what role it is playing.

A good public lead says: this unidentified person may match this missing person; here is the source; here is why; here is where the information should be sent. A bad public lead says: I solved it; here is the person's name; go look at their family.

A good investigator tries to disprove the favored theory. A bad one collects only the pieces that make the theory prettier.

A good true-crime production distinguishes what police established, what a witness said, what a document shows and what the creator believes. A bad one edits all four into the same confident voice.

The difference may sound fussy until an innocent person's name starts trending.

Crowds can be excellent at search. They are much less reliable at restraint. The modern ethical challenge is to preserve the first ability without pretending the second is optional.

What the Tupac verdict actually tells us

The Shakur verdict is an appropriate marker for this stage of true crime precisely because it resists the easy story.

For nearly 30 years, the killing existed in multiple systems at once: a police investigation, a hip-hop historical trauma, a conspiracy ecosystem, a documentary subject, an interview topic and a permanent internet mystery. Davis repeatedly spoke about the shooting in public. Those statements became part of the eventual prosecution.¹²

The trial ended Aug. 31 with a guilty verdict, but even that is not the same thing as total closure. Sentencing remains ahead. An appeal is expected. The identity of the actual shooter was not resolved at trial. The murder of Christopher Wallace remains unsolved. Shakur's family and millions of listeners have different definitions of what closure would mean.²

That is another lesson true crime has difficulty accepting: a legal ending and a narrative ending are not necessarily the same thing. The genre once promised audiences access to a mystery. The digital version quietly offered something more intoxicating, a role inside it.

Sometimes that role helps find a witness. Sometimes it gives an unidentified person back a name. Sometimes it keeps institutional failure visible long enough that a dormant case becomes active again. And sometimes it turns a stranger into a suspect because thousands of people agreed with one another before anybody checked.

The internet can keep a case alive. It can also keep a wound open.

The next evolution of true crime should not be figuring out how to make the audience investigate more. We already solved that technical problem. The harder problem is teaching a participatory audience that curiosity does not erase obligation - and that the most responsible sentence in any investigation may still be the one the genre likes least: we do not know yet.

Reporting & references

1. Reuters, “Jury finds ex-gang leader guilty of murder in Tupac killing case,” Sept. 1, 2026.
2. Associated Press, “Jury convicts a man of first-degree murder for the 1996 killing of rap icon Tupac Shakur,” Aug. 31, 2026; and AP follow-up on closure, Sept. 2, 2026.
3. Pew Research Center, “A Profile of the Top-Ranked Podcasts in the U.S.,” June 15, 2023.
4. Pew Research Center, “True crime podcasts are popular in the U.S.,” June 20, 2023.
5. Elizabeth Yardley et al., “What’s the deal with ‘websleuthing’?”, Crime, Media, Culture, 2018.
6. David Wästerfors, Veronika Burcar Alm and Erik Hannerz, “The bumpy paths of online sleuthing,” New Media & Society, 2024.
7. The Washington Post, “How Internet sleuths on Facebook and Reddit solved the 20-year-old mystery of a missing teenager,” Dec. 10, 2015.
8. The Doe Network, mission and operating principles. Source
9. NamUs, Frequently Asked Questions and current missing/unidentified-person statistics, updated 2026. Source
10. National Institute of Justice / Office of Justice Programs, NamUs Fiscal Year 2024 Annual Report, published April 2026.
11. Associated Press, reporting on the Kristin Smart investigation and the role of Your Own Backyard, including 2026 follow-up search coverage.
12. Los Angeles Times, “Boston bombing: Reddit learns how ‘witch hunts can start’,” April 20, 2013.
13. CBS News, “Reddit apologizes for Boston ‘online witch hunt’,” April 22, 2013.
14. Kelli S. Boling and Danielle C. Slakoff, “What an invasion, an immense invasion: Examining the adverse effects of true crime media on co-victims,” Crime, Media, Culture, 2025.
15. Society of Professional Journalists, Code of Ethics.
16. Society of Professional Journalists, proposed 2026 Code of Ethics revisions.
17. Christine Zozula, “Cold case comfort: Emotions, motivations, and identities of true crime fans,” Crime, Media, Culture, 2026.
18. Tomás Dodds et al., “On the Institutionalization of OSINV in Journalistic Practice,” Journalism & Mass Communication Quarterly, 2025.